This project is an extension of previous work involving the creation, development and testing of image-processing techniques designed to improve diagnostic performance. New algorithms for eliminating the effects of artifacts in subtraction radiographs have been devised. They involve position-invariant contrast manipulation, and context-dependent analysis of grey-level changes produced from sequentially obtained dental radiographs. Other work involves the use of image-processing techniques to enhance radiographic images produced digitally for use in tomosynthesis. One approach explores the use of spatial-frequency filtering techniques on x-ray projections prior to tomographic reconstruction. Another is based on an iteration scheme which eliminates artifacts produced by blurring of structures projected sharply in adjacent planes. Preliminary data suggest that these algorithms can eliminate artifacts which significantly compromise diagnostic performance obtainable from subtraction radiographs and tomosynthetic reconstructions.